October 21, 2019

3194 words 15 mins read

Paper Group AWR 25

Fast Maximum Likelihood estimation via Equilibrium Expectation for Large Network Data. Domain Adaptive Segmentation in Volume Electron Microscopy Imaging. Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression. FeTa: A DCA Pruning Algorithm with Generalization Error Guarantees. Ask No More: Deciding when to guess i …

Fast Maximum Likelihood estimation via Equilibrium Expectation for Large Network Data


Title	Fast Maximum Likelihood estimation via Equilibrium Expectation for Large Network Data
Authors	Maksym Byshkin, Alex Stivala, Antonietta Mira, Garry Robins, Alessandro Lomi
Abstract	A major line of contemporary research on complex networks is based on the development of statistical models that specify the local motifs associated with macro-structural properties observed in actual networks. This statistical approach becomes increasingly problematic as network size increases. In the context of current research on efficient estimation of models for large network data sets, we propose a fast algorithm for maximum likelihood estimation (MLE) that afords a signifcant increase in the size of networks amenable to direct empirical analysis. The algorithm we propose in this paper relies on properties of Markov chains at equilibrium, and for this reason it is called equilibrium expectation (EE). We demonstrate the performance of the EE algorithm in the context of exponential random graphmodels (ERGMs) a family of statistical models commonly used in empirical research based on network data observed at a single period in time. Thus far, the lack of efcient computational strategies has limited the empirical scope of ERGMs to relatively small networks with a few thousand nodes. The approach we propose allows a dramatic increase in the size of networks that may be analyzed using ERGMs. This is illustrated in an analysis of several biological networks and one social network with 104,103 nodes
Tasks
Published	2018-02-28
URL	http://arxiv.org/abs/1802.10311v2
PDF	http://arxiv.org/pdf/1802.10311v2.pdf
PWC	https://paperswithcode.com/paper/fast-maximum-likelihood-estimation-via
Repo	https://github.com/stivalaa/EstimNetDirected
Framework	none

Domain Adaptive Segmentation in Volume Electron Microscopy Imaging


Title	Domain Adaptive Segmentation in Volume Electron Microscopy Imaging
Authors	Joris Roels, Julian Hennies, Yvan Saeys, Wilfried Philips, Anna Kreshuk
Abstract	In the last years, automated segmentation has become a necessary tool for volume electron microscopy (EM) imaging. So far, the best performing techniques have been largely based on fully supervised encoder-decoder CNNs, requiring a substantial amount of annotated images. Domain Adaptation (DA) aims to alleviate the annotation burden by ‘adapting’ the networks trained on existing groundtruth data (source domain) to work on a different (target) domain with as little additional annotation as possible. Most DA research is focused on the classification task, whereas volume EM segmentation remains rather unexplored. In this work, we extend recently proposed classification DA techniques to an encoder-decoder layout and propose a novel method that adds a reconstruction decoder to the classical encoder-decoder segmentation in order to align source and target encoder features. The method has been validated on the task of segmenting mitochondria in EM volumes. We have performed DA from brain EM images to HeLa cells and from isotropic FIB/SEM volumes to anisotropic TEM volumes. In all cases, the proposed method has outperformed the extended classification DA techniques and the finetuning baseline. An implementation of our work can be found on https://github.com/JorisRoels/domain-adaptive-segmentation.
Tasks	Domain Adaptation
Published	2018-10-23
URL	http://arxiv.org/abs/1810.09734v2
PDF	http://arxiv.org/pdf/1810.09734v2.pdf
PWC	https://paperswithcode.com/paper/domain-adaptive-segmentation-in-volume
Repo	https://github.com/JorisRoels/domain-adaptive-segmentation
Framework	pytorch

Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression


Title	Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression
Authors	Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Fred Hohman, Siwei Li, Li Chen, Michael E. Kounavis, Duen Horng Chau
Abstract	The rapidly growing body of research in adversarial machine learning has demonstrated that deep neural networks (DNNs) are highly vulnerable to adversarially generated images. This underscores the urgent need for practical defense that can be readily deployed to combat attacks in real-time. Observing that many attack strategies aim to perturb image pixels in ways that are visually imperceptible, we place JPEG compression at the core of our proposed Shield defense framework, utilizing its capability to effectively “compress away” such pixel manipulation. To immunize a DNN model from artifacts introduced by compression, Shield “vaccinates” a model by re-training it with compressed images, where different compression levels are applied to generate multiple vaccinated models that are ultimately used together in an ensemble defense. On top of that, Shield adds an additional layer of protection by employing randomization at test time that compresses different regions of an image using random compression levels, making it harder for an adversary to estimate the transformation performed. This novel combination of vaccination, ensembling, and randomization makes Shield a fortified multi-pronged protection. We conducted extensive, large-scale experiments using the ImageNet dataset, and show that our approaches eliminate up to 94% of black-box attacks and 98% of gray-box attacks delivered by the recent, strongest attacks, such as Carlini-Wagner’s L2 and DeepFool. Our approaches are fast and work without requiring knowledge about the model.
Tasks
Published	2018-02-19
URL	http://arxiv.org/abs/1802.06816v1
PDF	http://arxiv.org/pdf/1802.06816v1.pdf
PWC	https://paperswithcode.com/paper/shield-fast-practical-defense-and-vaccination
Repo	https://github.com/rickyHong/JPEG-Defense-repl
Framework	tf

FeTa: A DCA Pruning Algorithm with Generalization Error Guarantees


Title	FeTa: A DCA Pruning Algorithm with Generalization Error Guarantees
Authors	Konstantinos Pitas, Mike Davies, Pierre Vandergheynst
Abstract	Recent DNN pruning algorithms have succeeded in reducing the number of parameters in fully connected layers, often with little or no drop in classification accuracy. However, most of the existing pruning schemes either have to be applied during training or require a costly retraining procedure after pruning to regain classification accuracy. We start by proposing a cheap pruning algorithm for fully connected DNN layers based on difference of convex functions (DC) optimisation, that requires little or no retraining. We then provide a theoretical analysis for the growth in the Generalization Error (GE) of a DNN for the case of bounded perturbations to the hidden layers, of which weight pruning is a special case. Our pruning method is orders of magnitude faster than competing approaches, while our theoretical analysis sheds light to previously observed problems in DNN pruning. Experiments on commnon feedforward neural networks validate our results.
Tasks
Published	2018-03-12
URL	http://arxiv.org/abs/1803.04239v1
PDF	http://arxiv.org/pdf/1803.04239v1.pdf
PWC	https://paperswithcode.com/paper/feta-a-dca-pruning-algorithm-with
Repo	https://github.com/konstantinos-p/FeTa_Fully_Connected
Framework	none

Ask No More: Deciding when to guess in referential visual dialogue


Title	Ask No More: Deciding when to guess in referential visual dialogue
Authors	Ravi Shekhar, Tim Baumgartner, Aashish Venkatesh, Elia Bruni, Raffaella Bernardi, Raquel Fernandez
Abstract	Our goal is to explore how the abilities brought in by a dialogue manager can be included in end-to-end visually grounded conversational agents. We make initial steps towards this general goal by augmenting a task-oriented visual dialogue model with a decision-making component that decides whether to ask a follow-up question to identify a target referent in an image, or to stop the conversation to make a guess. Our analyses show that adding a decision making component produces dialogues that are less repetitive and that include fewer unnecessary questions, thus potentially leading to more efficient and less unnatural interactions.
Tasks	Decision Making, Visual Dialog
Published	2018-05-17
URL	http://arxiv.org/abs/1805.06960v2
PDF	http://arxiv.org/pdf/1805.06960v2.pdf
PWC	https://paperswithcode.com/paper/ask-no-more-deciding-when-to-guess-in
Repo	https://github.com/shekharRavi/ask-no-more-COLING2018
Framework	pytorch

Boundary loss for highly unbalanced segmentation


Title	Boundary loss for highly unbalanced segmentation
Authors	Hoel Kervadec, Jihene Bouchtiba, Christian Desrosiers, Éric Granger, Jose Dolz, Ismail Ben Ayed
Abstract	Widely used loss functions for convolutional neural network (CNN) segmentation, e.g., Dice or cross-entropy, are based on integrals (summations) over the segmentation regions. Unfortunately, it is quite common in medical image analysis to have highly unbalanced segmentations, where standard losses contain regional terms with values that differ considerably –typically of several orders of magnitude– across segmentation classes, which may affect training performance and stability. The purpose of this study is to build a boundary loss, which takes the form of a distance metric on the space of contours, not regions. We argue that a boundary loss can mitigate the difficulties of regional losses in the context of highly unbalanced segmentation problems because it uses integrals over the boundary between regions instead of unbalanced integrals over regions. Furthermore, a boundary loss provides information that is complementary to regional losses. Unfortunately, it is not straightforward to represent the boundary points corresponding to the regional softmax outputs of a CNN. Our boundary loss is inspired by discrete (graph-based) optimization techniques for computing gradient flows of curve evolution. Following an integral approach for computing boundary variations, we express a non-symmetric L2 distance on the space of shapes as a regional integral, which avoids completely local differential computations involving contour points. Our boundary loss is the sum of linear functions of the regional softmax probability outputs of the network. Therefore, it can easily be combined with standard regional losses and implemented with any existing deep network architecture for N-D segmentation. Our boundary loss has been validated on two benchmark datasets corresponding to difficult, highly unbalanced segmentation problems: the ischemic stroke lesion (ISLES) and white matter hyperintensities (WMH).
Tasks	Brain Lesion Segmentation From Mri, Ischemic Stroke Lesion Segmentation, Lesion Segmentation, Medical Image Segmentation, Semantic Segmentation, Unbalanced Segmentation
Published	2018-12-17
URL	https://arxiv.org/abs/1812.07032v2
PDF	https://arxiv.org/pdf/1812.07032v2.pdf
PWC	https://paperswithcode.com/paper/boundary-loss-for-highly-unbalanced
Repo	https://github.com/LIVIAETS/surface-loss
Framework	pytorch

Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing


Title	Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing
Authors	Magnus Wrenninge, Jonas Unger
Abstract	We introduce Synscapes – a synthetic dataset for street scene parsing created using photorealistic rendering techniques, and show state-of-the-art results for training and validation as well as new types of analysis. We study the behavior of networks trained on real data when performing inference on synthetic data: a key factor in determining the equivalence of simulation environments. We also compare the behavior of networks trained on synthetic data and evaluated on real-world data. Additionally, by analyzing pre-trained, existing segmentation and detection models, we illustrate how uncorrelated images along with a detailed set of annotations open up new avenues for analysis of computer vision systems, providing fine-grain information about how a model’s performance changes according to factors such as distance, occlusion and relative object orientation.
Tasks	Scene Parsing, Street Scene Parsing
Published	2018-10-19
URL	http://arxiv.org/abs/1810.08705v1
PDF	http://arxiv.org/pdf/1810.08705v1.pdf
PWC	https://paperswithcode.com/paper/synscapes-a-photorealistic-synthetic-dataset
Repo	https://github.com/MartinHahner88/FoggySynscapes
Framework	none

Overview of CAIL2018: Legal Judgment Prediction Competition


Title	Overview of CAIL2018: Legal Judgment Prediction Competition
Authors	Haoxi Zhong, Chaojun Xiao, Zhipeng Guo, Cunchao Tu, Zhiyuan Liu, Maosong Sun, Yansong Feng, Xianpei Han, Zhen Hu, Heng Wang, Jianfeng Xu
Abstract	In this paper, we give an overview of the Legal Judgment Prediction (LJP) competition at Chinese AI and Law challenge (CAIL2018). This competition focuses on LJP which aims to predict the judgment results according to the given facts. Specifically, in CAIL2018 , we proposed three subtasks of LJP for the contestants, i.e., predicting relevant law articles, charges and prison terms given the fact descriptions. CAIL2018 has attracted several hundreds participants (601 teams, 1, 144 contestants from 269 organizations). In this paper, we provide a detailed overview of the task definition, related works, outstanding methods and competition results in CAIL2018.
Tasks
Published	2018-10-13
URL	http://arxiv.org/abs/1810.05851v1
PDF	http://arxiv.org/pdf/1810.05851v1.pdf
PWC	https://paperswithcode.com/paper/overview-of-cail2018-legal-judgment
Repo	https://github.com/JepsonWong/Text_Classification
Framework	tf

Layer rotation: a surprisingly powerful indicator of generalization in deep networks?


Title	Layer rotation: a surprisingly powerful indicator of generalization in deep networks?
Authors	Simon Carbonnelle, Christophe De Vleeschouwer
Abstract	Our work presents extensive empirical evidence that layer rotation, i.e. the evolution across training of the cosine distance between each layer’s weight vector and its initialization, constitutes an impressively consistent indicator of generalization performance. In particular, larger cosine distances between final and initial weights of each layer consistently translate into better generalization performance of the final model. Interestingly, this relation admits a network independent optimum: training procedures during which all layers’ weights reach a cosine distance of 1 from their initialization consistently outperform other configurations -by up to 30% test accuracy. Moreover, we show that layer rotations are easily monitored and controlled (helpful for hyperparameter tuning) and potentially provide a unified framework to explain the impact of learning rate tuning, weight decay, learning rate warmups and adaptive gradient methods on generalization and training speed. In an attempt to explain the surprising properties of layer rotation, we show on a 1-layer MLP trained on MNIST that layer rotation correlates with the degree to which features of intermediate layers have been trained.
Tasks
Published	2018-06-05
URL	https://arxiv.org/abs/1806.01603v2
PDF	https://arxiv.org/pdf/1806.01603v2.pdf
PWC	https://paperswithcode.com/paper/on-layer-level-control-of-dnn-training-and
Repo	https://github.com/ispgroupucl/layer-rotation-paper-experiments
Framework	tf

Sampling Using Neural Networks for colorizing the grayscale images


Title	Sampling Using Neural Networks for colorizing the grayscale images
Authors	Wonbong Jang
Abstract	The main idea of this paper is to explore the possibilities of generating samples from the neural networks, mostly focusing on the colorization of the grey-scale images. I will compare the existing methods for colorization and explore the possibilities of using new generative modeling to the task of colorization. The contributions of this paper are to compare the existing structures with similar generating structures(Decoders) and to apply the novel structures including Conditional VAE(CVAE), Conditional Wasserstein GAN with Gradient Penalty(CWGAN-GP), CWGAN-GP with L1 reconstruction loss, Adversarial Generative Encoders(AGE) and Introspective VAE(IVAE). I trained these models using CIFAR-10 images. To measure the performance, I use Inception Score(IS) which measures how distinctive each image is and how diverse overall samples are as well as human eyes for CIFAR-10 images. It turns out that CVAE with L1 reconstruction loss and IVAE achieve the highest score in IS. CWGAN-GP with L1 tends to learn faster than CWGAN-GP, but IS does not increase from CWGAN-GP. CWGAN-GP tends to generate more diverse images than other models using reconstruction loss. Also, I figured out that the proper regularization plays a vital role in generative modeling.
Tasks	Colorization
Published	2018-12-27
URL	http://arxiv.org/abs/1812.10650v1
PDF	http://arxiv.org/pdf/1812.10650v1.pdf
PWC	https://paperswithcode.com/paper/sampling-using-neural-networks-for-colorizing
Repo	https://github.com/wayne1123/colorization
Framework	pytorch

Neighbourhood Consensus Networks


Title	Neighbourhood Consensus Networks
Authors	Ignacio Rocco, Mircea Cimpoi, Relja Arandjelović, Akihiko Torii, Tomas Pajdla, Josef Sivic
Abstract	We address the problem of finding reliable dense correspondences between a pair of images. This is a challenging task due to strong appearance differences between the corresponding scene elements and ambiguities generated by repetitive patterns. The contributions of this work are threefold. First, inspired by the classic idea of disambiguating feature matches using semi-local constraints, we develop an end-to-end trainable convolutional neural network architecture that identifies sets of spatially consistent matches by analyzing neighbourhood consensus patterns in the 4D space of all possible correspondences between a pair of images without the need for a global geometric model. Second, we demonstrate that the model can be trained effectively from weak supervision in the form of matching and non-matching image pairs without the need for costly manual annotation of point to point correspondences. Third, we show the proposed neighbourhood consensus network can be applied to a range of matching tasks including both category- and instance-level matching, obtaining the state-of-the-art results on the PF Pascal dataset and the InLoc indoor visual localization benchmark.
Tasks	Visual Localization
Published	2018-10-24
URL	http://arxiv.org/abs/1810.10510v2
PDF	http://arxiv.org/pdf/1810.10510v2.pdf
PWC	https://paperswithcode.com/paper/neighbourhood-consensus-networks
Repo	https://github.com/XiSHEN0220/NeighConsensus
Framework	pytorch

GestureGAN for Hand Gesture-to-Gesture Translation in the Wild


Title	GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
Authors	Hao Tang, Wei Wang, Dan Xu, Yan Yan, Nicu Sebe
Abstract	Hand gesture-to-gesture translation in the wild is a challenging task since hand gestures can have arbitrary poses, sizes, locations and self-occlusions. Therefore, this task requires a high-level understanding of the mapping between the input source gesture and the output target gesture. To tackle this problem, we propose a novel hand Gesture Generative Adversarial Network (GestureGAN). GestureGAN consists of a single generator $G$ and a discriminator $D$, which takes as input a conditional hand image and a target hand skeleton image. GestureGAN utilizes the hand skeleton information explicitly, and learns the gesture-to-gesture mapping through two novel losses, the color loss and the cycle-consistency loss. The proposed color loss handles the issue of “channel pollution” while back-propagating the gradients. In addition, we present the Fr'echet ResNet Distance (FRD) to evaluate the quality of generated images. Extensive experiments on two widely used benchmark datasets demonstrate that the proposed GestureGAN achieves state-of-the-art performance on the unconstrained hand gesture-to-gesture translation task. Meanwhile, the generated images are in high-quality and are photo-realistic, allowing them to be used as data augmentation to improve the performance of a hand gesture classifier. Our model and code are available at https://github.com/Ha0Tang/GestureGAN.
Tasks	Data Augmentation, Gesture-to-Gesture Translation
Published	2018-08-14
URL	https://arxiv.org/abs/1808.04859v2
PDF	https://arxiv.org/pdf/1808.04859v2.pdf
PWC	https://paperswithcode.com/paper/gesturegan-for-hand-gesture-to-gesture
Repo	https://github.com/Ha0Tang/GestureGAN
Framework	pytorch

Mix and match networks: encoder-decoder alignment for zero-pair image translation


Title	Mix and match networks: encoder-decoder alignment for zero-pair image translation
Authors	Yaxing Wang, Joost van de Weijer, Luis Herranz
Abstract	We address the problem of image translation between domains or modalities for which no direct paired data is available (i.e. zero-pair translation). We propose mix and match networks, based on multiple encoders and decoders aligned in such a way that other encoder-decoder pairs can be composed at test time to perform unseen image translation tasks between domains or modalities for which explicit paired samples were not seen during training. We study the impact of autoencoders, side information and losses in improving the alignment and transferability of trained pairwise translation models to unseen translations. We show our approach is scalable and can perform colorization and style transfer between unseen combinations of domains. We evaluate our system in a challenging cross-modal setting where semantic segmentation is estimated from depth images, without explicit access to any depth-semantic segmentation training pairs. Our model outperforms baselines based on pix2pix and CycleGAN models.
Tasks	Colorization, Semantic Segmentation, Style Transfer
Published	2018-04-06
URL	http://arxiv.org/abs/1804.02199v1
PDF	http://arxiv.org/pdf/1804.02199v1.pdf
PWC	https://paperswithcode.com/paper/mix-and-match-networks-encoder-decoder
Repo	https://github.com/yaxingwang/Mix-and-match-networks
Framework	tf

Deep Learning for Sentiment Analysis : A Survey


Title	Deep Learning for Sentiment Analysis : A Survey
Authors	Lei Zhang, Shuai Wang, Bing Liu
Abstract	Deep learning has emerged as a powerful machine learning technique that learns multiple layers of representations or features of the data and produces state-of-the-art prediction results. Along with the success of deep learning in many other application domains, deep learning is also popularly used in sentiment analysis in recent years. This paper first gives an overview of deep learning and then provides a comprehensive survey of its current applications in sentiment analysis.
Tasks	Sentiment Analysis
Published	2018-01-24
URL	http://arxiv.org/abs/1801.07883v2
PDF	http://arxiv.org/pdf/1801.07883v2.pdf
PWC	https://paperswithcode.com/paper/deep-learning-for-sentiment-analysis-a-survey
Repo	https://github.com/ridakadri14/AspectBasedSentimentAnalysis
Framework	tf

SeqFace: Make full use of sequence information for face recognition


Title	SeqFace: Make full use of sequence information for face recognition
Authors	Wei Hu, Yangyu Huang, Fan Zhang, Ruirui Li, Wei Li, Guodong Yuan
Abstract	Deep convolutional neural networks (CNNs) have greatly improved the Face Recognition (FR) performance in recent years. Almost all CNNs in FR are trained on the carefully labeled datasets containing plenty of identities. However, such high-quality datasets are very expensive to collect, which restricts many researchers to achieve state-of-the-art performance. In this paper, we propose a framework, called SeqFace, for learning discriminative face features. Besides a traditional identity training dataset, the designed SeqFace can train CNNs by using an additional dataset which includes a large number of face sequences collected from videos. Moreover, the label smoothing regularization (LSR) and a new proposed discriminative sequence agent (DSA) loss are employed to enhance discrimination power of deep face features via making full use of the sequence data. Our method achieves excellent performance on Labeled Faces in the Wild (LFW), YouTube Faces (YTF), only with a single ResNet. The code and models are publicly available on-line (https://github.com/huangyangyu/SeqFace).
Tasks	Face Recognition, Face Verification
Published	2018-03-17
URL	http://arxiv.org/abs/1803.06524v2
PDF	http://arxiv.org/pdf/1803.06524v2.pdf
PWC	https://paperswithcode.com/paper/seqface-make-full-use-of-sequence-information
Repo	https://github.com/huangyangyu/SeqFace
Framework	none